mistralai
Mistral: Voxtral Small 24B 2507 STT
mistralai/voxtral-small-24b-2507-stt
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
Mistral's Voxtral Small 24B 2507 STT is listed in the OpenRouter catalog as a speech-to-text variant of Voxtral Small that accepts audio input and produces transcription output. The underlying Voxtral Small model card describes a 24B audio-language model built on Mistral Small 3, offering automatic language detection across eight languages, a 32k token context handling roughly 30 minutes of transcription audio, and modes for transcription, audio question answering, summarization, and experimental function calling. Benchmark figures are vendor-reported; no extractable scores appear in this source pack.[1][2]
Strengths
- The model card describes a dedicated transcription mode with automatic source-language detection across eight widely used languages, including English, Spanish, French, and Portuguese.[2]
- Per the model card, a 32k token context handles up to about 30 minutes of audio for transcription or 40 minutes for audio understanding, suiting long-form material.[2]
- The underlying model reportedly supports audio Q&A, summarization, and experimental function calling, though this catalog variant is listed with transcription-only output.[1][2]
Limitations
- Benchmark results in the model card are image-only, so no verifiable scores exist here; OpenRouter lists no benchmark data. Validate accuracy on your own audio before adoption.[1][2]
- The model card notes system prompts are not yet supported and function calling is experimental; running the underlying model on GPU reportedly requires about 55 GB of VRAM in bf16 or fp16.[2]
- The -stt variant has no independent model card in this pack, so variant-specific behavior beyond the catalog's transcription-output listing is unverified.[1]
Where to use it
- Multilingual speech-to-text pipelines may benefit, given the reported automatic language detection across eight languages; confirm coverage and accuracy for target languages.[2]
- Long-form transcription of meetings, interviews, or lectures could fit, as the model card claims roughly 30 minutes of audio per request.[2]
- Workloads that need a single audio model for transcription plus downstream understanding might consider the broader Voxtral family, though this variant is cataloged for transcription output only.[1][2]
Sources
- [1] OpenRouter catalog · Mistral: Voxtral Small 24B 2507 STTSources checked: Nov 13, 2026
- [2] Model documentation · Mistral: Voxtral Small 24B 2507 STTSources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Voxtral Small 24B 2507 STT is a speech transcription model from Mistral AI. It is suited for transcription, translation, and audio understanding workloads that benefit from its larger model capacity.
Specifications
- Model ID
- mistralai/voxtral-small-24b-2507-stt
- Canonical ID
- mistralai/voxtral-small-24b-2507-stt-20260813
- Creator
- mistralai
- Added to OpenRouter
- 13.08.2026
- Context length
- 0
- Modality
- audio->transcription
- Input modalities
- audio
- Output modalities
- transcription
- Token pricing
- No per-token price in the source snapshot — see OpenRouter for the billing unit
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | — |
| Coding | No measurement | — |
| Agents | No measurement | — |
| Writing | No measurement | — |
| Reasoning | No measurement | — |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | input: audio |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
No evaluations are supplied for this model.
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| DeepInfrabf16 | 0 | — | — | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.